Under Red Hat at least, as long as your video card is supported by the distro (and you do check, right), then it would be a matter of running Xconfigurator. It's in the docs that come with the box set as well as at www.redhat.com. I know mandrake and suse also have tools to configure X after the initial installation. If we're talking about a video card that isn't supported by the version of X that comes with your distro, to bad, it's not gonna be easy to get it to work. Not that it would be easy to get a card with no OS X drivers to work under OS X though. With wireless cards it's the same deal. If it's supported it should just work. If it's not, expect to work. It's like that on every other OS.
OS X has lots of config files. Havent you ever opened the Terminal.app and poked around? Most applications have a xml based global config file and user local config file./etc is full of your normal unix config files (tho the syntax is different on many of them). And dont get me started on Netinfo. Netinfo makes doing the simplest things into a giant headache. It's sorta like the windows registry, and IMHO, sucks just as bad.
The locations of libraries in Linux is standardized. Anyone not following the standard should be flogged. Furthermore, the Linux dynamic linker is capable of finding the library as long as it's in a directory it knows about. So, unless a distro has a lib in a non-standard location *and* has a misconfigured dynamic linker, the application writer doesn't need to know or care about where the lib is on the target system.
RPMS are a very nice and easy way to install something. The fact that SuSE (the biggest culprit) handles RPMS differently then just about everyone else doesn't mean they're non standard. And I specifically said, make 2 or 3 RPMS for your common desktop distros and a tarball for everyone else. I'm not gonna complain if you give me a tarball to install. "The average user" probably isn't who's gonna be installing quicktime on linux anyway which makes the entire point moot.
Window managers are meaningless in an applications context. In linux you don't care what window manager is running. The application doesn't even know the window manager exists. X has been backwards compatible for at least 10 years. And the distros did get together and decide where things were gonna be. All directory locations are standardized now. Not every distro strickly adhears to it, but as in my previous point, most of those locations are not something you need to worry about. A standard desktop isn't necessary either. Especially if we're talking about quicktime which happens to break most of Apple's UI guidelines, who cares if it looks out of place on linux too.
And, in conclusion, as I've stated several times already, *there is a standard for what files go where*. Google for the FHS. It will tell you all about it. Linux works fine, lots of commercial applications work on it just fine. There is no reason apple couldn't release a quicktime player for Linux and the all of the arguments you give are either unnecessary details (the window manager), moot points (tarballs vs rpms... linux users would be happy with anything, and the average user isn't using linux), or nitpicking the linux desktop (standard UI's? maybe if Quicktime used the MacOS stardard UI you'd have a point), etc.
That was one of the biggest loads of crap I've ever read. Lets start from the bottom and work our way up. OS X has lots of text config files and lots of GUI config programs sitting on top of them. I'm using Red Hat 7.2 and I don't need to touch config files unless I want to (this is for all desktop style configuration, when it comes to servers, YMMV). Lets all try using the Linux distros of 2002 and forget about the Linux distros of 1998 (unless you're still using debian stable, because then you're still living the linux distro of 1998... *zing*).
Next, that whole "we need to test under 2^64 different configurations" argument is crap. Provide an RPM for Red Hat and maybe Mandrake and a tarball. If you require Gtk or Qt or wxWindows or whatever, say so. The Linux kernel is the Linux kernel. Unless you are using wierd ioctls or installing a device driver (and even for that, NVidia and VMWare are managing)there is nothing to test. X, same deal, it's been binary compatible for 10 years, there was even a slashdot article about it. Real has managed to release a Linux version of their media player for years and I've never heard people complain about it not being available for everyone. I mean, everyone on every platform complains that real sucks, but no one complains about availablity.
See, the big problem in my opinion is all these software development houses are so used to developing for Windows or the Mac where changing API's and system calls out from underneath everyone that they can't think of things being different. The Unix system call set hasn't changed significantly in probably 15 years. X hasn't changed in at least 10. Gtk and Qt change, but all point releases stay compatible and no one is gonna scream if you say you require Gtk 1.2 or Qt 2.x. There is no "business reason" to screw your users in an open source development model so no one does so.
In conclusion, if you want to post on why a company should or shouldn't do port something to Linux, you should have at least a little knowledge of what Linux in 2002 is like or you come off sounding a) ignorant or b) a troll.
Please, stop comparing RPMS to apt-get. Compare rpm to dpkg all you want, but apt-get is completely different. If you want apt-get functionality on Red Hat use up2date, if you want it on Mandrake use urpmi. These tools have existed in their respective distributions for several releases now and work just as well as apt-get.
I'm sure when reiser becomes stable they will support it. Same with xfs and jfs and lvm. But, since that day has not yet happened, I can promise you that reiser won't be in 7.3.
No, Navigator was free only for educational use and some forms of personal use. Netscape made most of their money selling the browser to OEM's for preloads and companies. Navigator was not free for everyone until after Microsoft "integrated" IE into Windows.
They weren't sending malformed headers at all. They sent FROM, RCPT, and QUIT. If your mailserver manages to crash by being sent those headers (no matter what the parameters to those headers are) the problem is not with the email that crashed the server.
Nope, you're actually wrong. Red Hat Linux is an OS. It is an OS made up of the Linux kernel, and a mixture of GNU, Red Hat, and other OSS userland tools. SuSE, Debian, Mandrake, etc. are all also different flavors of a Linux based OS which integrate GNU, OSS and their own tools to build their own OS.
Not quite true. They try and show everything subbed, but in some cases they can't. Case in point, they aired Serial Experiment Lain complete and uncut, but the publisher only provided them with dubbed versions so they had no option.
The X Clipboard and XDnD are both defined protocols which, unfortunately, Qt2 (and thus KDE2) did not follow. Qt3 does, so hopefully, the clipboard and dnd between Gtk and Qt apps will work correctly.
Microsoft software really doesn't have significantly more problems than any other software. Microsoft is simply a large target, and so many and more people spend much more time finding those holes (often for malicious purposes, sadly).
IE has the biggest marketshare, and Windows has the biggest desktop marketshare, but the reason that people attack Windows systems is it's easy. I wish people would stop kidding themselves with the market share excuse. MS software has serious design flaws which makes it very easy to exploit a flaw in the browser to extract data from the registry and mail that off to some email address. Under windows, that is easy, under Linux there are multiple different browsers, you don't know what email client might be available, there is no central place to grab system/user info and there is no easy way to automate the process. The same type of exploit is used over and over and over again, yet for every patch MS releases, someone finds a new way to write an exploit that uses the same basic method. How long, exactly, do you think it's going to take before Microsoft recognizes this and fixes the design flaws instead of releasing patches which amount to little more then sticking their finger in the crack in the dam?
Where exactly in my post did I bash Windows? I pointed out that I agree that server admins that don't patch systems are at fault if their server has problems. But, I also pointed out that we're not talking about servers being exploited, we're talking about software which runs on top of Windows consistently have gaping holes.
Microsoft could put as many firewalls as they want to into their product, but IE (or out look, or uPNP, name your exploit of the week) continually has holes in it. A web browser should not have the capability to cause the havoc that IE has caused for Microsoft users (and don't even get me started on Outlook).
Not that there isn't plenty of reasons to bash Windows, please don't imply that I'm doing something that I'm obviously not doing (and don't automatically assume that I know nothing about Windows just because I know something of Linux).
Hmm... or maybe it's because that problem with Linux went away a long time ago. A default workstation install of Red Hat Linux 7.2 has zero open ports and a firewall that blocks access to all ports under 1024.
Now, obviously if someone sets up a server and doesn't patch, that person is an idiot (and that is true no matter what OS he/she is running). Unfortunately for your argument, we're talking about an instant messenger client and a web browser, not things that are likely to be installed on a server. The fact is, you can't exploit my Linux system via Mozilla/Konq/Galeon/Netscape, yet every other week, a new way to exploit Windows using IE pops up.
So, in conclusion, your argument is completely irrelevant to the topic at hand... there has never been an exploit like this released against Linux, there is an exploit like this released against Windows about once a month... I think we can safely start saying it's Microsoft's fault at some point.
What stores do you shop in? Fry's has (on the shelves) Windows 2000, 2000 Pro, 2000 Server, XP Home, XP Pro, and Me. Best Buy and Circuit City have mostly the same (they didn't have 2000 Server IIRC). Even if you don't want to count the different flavors of the same version, that is 3 different versions that are easily available at the store.
And outside of stores, there are all of the different embedded versions of Windows, etc. MS has had one platform for the past 7 years... it was called Win32. The fact that they broke the API repeatedly, and every different patch level of every different version of Windows caused new and different bugs in Win32 does not make me feel even the slightest bit confident that XP is going to make this any better.
And XP is supposed to be phased out within 2 years for the next version of Windows that will be all.NET (another new platform). MS will make sure that you always need to have OS specific hooks and code paths to work reliably across the commonly used versions of Windows.
If people actually went to the Mono website and read some things about it they can see that *THE MONO TEAM DOESN'T CARE IF MICROSOFT FOLLOWS THE SPEC*
C#, the CLR, and the Class Libraries *could* solve some of the problems that Miguel saw Gnome was having. Since it could solve some of these problems they are implementing the ECMA spec. If MS doesn't follow the spec, great. If they do follow the spec, great. This *does not effect Mono at all*. The project is not looking for interoperability. If they get it, it's a great benefit, but the primary goal is to solve problems that they personally have.
If you'd have read the article you'd have seen that it comes with a 40GB harddisk (for the internal bay) and 100 Mb/s ethernet adaptor, a usb keyboard and mouse, a vga output adaptor *and* the software. I don't know about you, but that looks like it could be about $200 worth of hardware alone if it's quality stuff.
So, you work on parts of Gnome, but API references and thousands of lines of code using those API's isn't enough for you? I'm sorry, but what parts of Gnome do you work on? I've worked on parts of Gnome too, and yet, somehow, I've managed. Hell, back in the Gnome 0.30 days, there was no documentation to speak of at all. And if you're complaining, why aren't you complaining about extensive documentation about libraries like Bonobo and GConf? They're certainly used by more applications then eel and gtkhtml (gtkhtml2 is a bonobo component to boot). Even some docs on just writing Nautilus Views would be alot more useful then docs on eel or librsvg.
Btw, before claiming someone should get a clue and information, make sure they don't know what they're talking about.
libeel is private to Nautilus. The other three are documented if you download them from cvs and run gtk-doc on them. And, none of those are core gnome libraries so you still haven't said anything about gnome core documentation.
As far as Nautilus and gnumeric, those are huge apps and porting will take a while. Of course they aren't adding features. But the code is being cleaned up as they go.
Everyone is. The day of the IPO, the stock opens on the market at that price. There have been IPO's where the stock went down on the first day. The fact that someone doesn't manage to buy the stock at that price goes up is just too bad. That's how the stock market works. If everyone got to preorder IPO shares at that price, then the price would never move on the stock, thus defeating the purpose of the stock market.
Tell me what it is you're referring about and perhaps I can respond. As far as I've seen, the biggest difference between KDE 2.2 and 3 is they're using Qt 3 now. And I can't really say that Qt 3 impresses me very much. OK, they added database aware widgets, I'll use Gnome-db.
So because gnome hasn't up'd it's version number fast enought it's not as good as KDE? (I remember everyone complaining about them bumping it up too fast back around the 1.0 release).
Gnome 2 is internationalized, has antialiased text, has a very configurable interface. The control center has been just about completely rewritten and is very slick.
If there is a UI difference between apps, complain to the app writer. But, gtk2 will make it much easier to write apps with a common look and feel and has made some nice improvements to the theme system.
Kmail is nice, Evolution is nicer, IMHO and Pan is just as good or better than KNode (again, IMHO). Glade and libglade couldn't make writing apps easier and Anjuta (especially with the work they're doing on Anjuta 2) is a very nice IDE.
If you want to think you are that much better than me for using KDE, please go ahead and do so. But your comment shows that you really are not aware of the capabilities of the current Gnome or of the huge advances that Gnome 2 has made. Things like the Pango font render, Bonobo, etc. are at the cutting edge of Linux desktop technology.
1. So go download one. There are Linux kernel debuggers available, just none in the default kernel tree.
2. Grub
3. You'll always find flaws in your design no matter how much thought is put into it. Thus the evolutionary progress of *all* OS's. OS X versus OS X.1 is a great example of that.
4. I would have to say that that number of people using and developing Linux under all the different configurations out there can be construed as taking input from "a number of external sources." Not to mention Red Hat, Mandrake, SuSE, etc. and their feedback.
5. The Linux source code is controlled by the maintainer of the kernel series. Different people have tried different source control tools and none have worked yet.
I'm not going to say there aren't things wrong with Linux, but you certainly haven't pointed any out here. And none of what you said makes it easier to debug OS X. And even if you do find a bug in it, are you going to write up a patch and submit it to the maintainers? Nope, didn't think so.
Slashdot Mirror
Under Red Hat at least, as long as your video card is supported by the distro (and you do check, right), then it would be a matter of running Xconfigurator. It's in the docs that come with the box set as well as at www.redhat.com. I know mandrake and suse also have tools to configure X after the initial installation. If we're talking about a video card that isn't supported by the version of X that comes with your distro, to bad, it's not gonna be easy to get it to work. Not that it would be easy to get a card with no OS X drivers to work under OS X though. With wireless cards it's the same deal. If it's supported it should just work. If it's not, expect to work. It's like that on every other OS.
/etc is full of your normal unix config files (tho the syntax is different on many of them). And dont get me started on Netinfo. Netinfo makes doing the simplest things into a giant headache. It's sorta like the windows registry, and IMHO, sucks just as bad.
... linux users would be happy with anything, and the average user isn't using linux), or nitpicking the linux desktop (standard UI's? maybe if Quicktime used the MacOS stardard UI you'd have a point), etc.
OS X has lots of config files. Havent you ever opened the Terminal.app and poked around? Most applications have a xml based global config file and user local config file.
The locations of libraries in Linux is standardized. Anyone not following the standard should be flogged. Furthermore, the Linux dynamic linker is capable of finding the library as long as it's in a directory it knows about. So, unless a distro has a lib in a non-standard location *and* has a misconfigured dynamic linker, the application writer doesn't need to know or care about where the lib is on the target system.
RPMS are a very nice and easy way to install something. The fact that SuSE (the biggest culprit) handles RPMS differently then just about everyone else doesn't mean they're non standard. And I specifically said, make 2 or 3 RPMS for your common desktop distros and a tarball for everyone else. I'm not gonna complain if you give me a tarball to install. "The average user" probably isn't who's gonna be installing quicktime on linux anyway which makes the entire point moot.
Window managers are meaningless in an applications context. In linux you don't care what window manager is running. The application doesn't even know the window manager exists. X has been backwards compatible for at least 10 years. And the distros did get together and decide where things were gonna be. All directory locations are standardized now. Not every distro strickly adhears to it, but as in my previous point, most of those locations are not something you need to worry about. A standard desktop isn't necessary either. Especially if we're talking about quicktime which happens to break most of Apple's UI guidelines, who cares if it looks out of place on linux too.
And, in conclusion, as I've stated several times already, *there is a standard for what files go where*. Google for the FHS. It will tell you all about it. Linux works fine, lots of commercial applications work on it just fine. There is no reason apple couldn't release a quicktime player for Linux and the all of the arguments you give are either unnecessary details (the window manager), moot points (tarballs vs rpms
That was one of the biggest loads of crap I've ever read. Lets start from the bottom and work our way up. OS X has lots of text config files and lots of GUI config programs sitting on top of them. I'm using Red Hat 7.2 and I don't need to touch config files unless I want to (this is for all desktop style configuration, when it comes to servers, YMMV). Lets all try using the Linux distros of 2002 and forget about the Linux distros of 1998 (unless you're still using debian stable, because then you're still living the linux distro of 1998 ... *zing*).
Next, that whole "we need to test under 2^64 different configurations" argument is crap. Provide an RPM for Red Hat and maybe Mandrake and a tarball. If you require Gtk or Qt or wxWindows or whatever, say so. The Linux kernel is the Linux kernel. Unless you are using wierd ioctls or installing a device driver (and even for that, NVidia and VMWare are managing)there is nothing to test. X, same deal, it's been binary compatible for 10 years, there was even a slashdot article about it. Real has managed to release a Linux version of their media player for years and I've never heard people complain about it not being available for everyone. I mean, everyone on every platform complains that real sucks, but no one complains about availablity.
See, the big problem in my opinion is all these software development houses are so used to developing for Windows or the Mac where changing API's and system calls out from underneath everyone that they can't think of things being different. The Unix system call set hasn't changed significantly in probably 15 years. X hasn't changed in at least 10. Gtk and Qt change, but all point releases stay compatible and no one is gonna scream if you say you require Gtk 1.2 or Qt 2.x. There is no "business reason" to screw your users in an open source development model so no one does so.
In conclusion, if you want to post on why a company should or shouldn't do port something to Linux, you should have at least a little knowledge of what Linux in 2002 is like or you come off sounding a) ignorant or b) a troll.
Please, stop comparing RPMS to apt-get. Compare rpm to dpkg all you want, but apt-get is completely different. If you want apt-get functionality on Red Hat use up2date, if you want it on Mandrake use urpmi. These tools have existed in their respective distributions for several releases now and work just as well as apt-get.
I'm sure when reiser becomes stable they will support it. Same with xfs and jfs and lvm. But, since that day has not yet happened, I can promise you that reiser won't be in 7.3.
In case your interested in Kite, the original version has been released in the US, uncut.
He didn't mark the PS2 DVD accessory as costing money because it's optional. The DVD capabilities of the PS2 work just fine without the remote.
No, Navigator was free only for educational use and some forms of personal use. Netscape made most of their money selling the browser to OEM's for preloads and companies. Navigator was not free for everyone until after Microsoft "integrated" IE into Windows.
They weren't sending malformed headers at all. They sent FROM, RCPT, and QUIT. If your mailserver manages to crash by being sent those headers (no matter what the parameters to those headers are) the problem is not with the email that crashed the server.
Nope, you're actually wrong. Red Hat Linux is an OS. It is an OS made up of the Linux kernel, and a mixture of GNU, Red Hat, and other OSS userland tools. SuSE, Debian, Mandrake, etc. are all also different flavors of a Linux based OS which integrate GNU, OSS and their own tools to build their own OS.
Not quite true. They try and show everything subbed, but in some cases they can't. Case in point, they aired Serial Experiment Lain complete and uncut, but the publisher only provided them with dubbed versions so they had no option.
That is a GSM phone. In the US, Cingular on the west coast and VoiceStream on the east coast both offer GSM service.
The X Clipboard and XDnD are both defined protocols which, unfortunately, Qt2 (and thus KDE2) did not follow. Qt3 does, so hopefully, the clipboard and dnd between Gtk and Qt apps will work correctly.
Microsoft software really doesn't have significantly more problems than any other software. Microsoft is simply a large target, and so many and more people spend much more time finding those holes (often for malicious purposes, sadly).
IE has the biggest marketshare, and Windows has the biggest desktop marketshare, but the reason that people attack Windows systems is it's easy. I wish people would stop kidding themselves with the market share excuse. MS software has serious design flaws which makes it very easy to exploit a flaw in the browser to extract data from the registry and mail that off to some email address. Under windows, that is easy, under Linux there are multiple different browsers, you don't know what email client might be available, there is no central place to grab system/user info and there is no easy way to automate the process. The same type of exploit is used over and over and over again, yet for every patch MS releases, someone finds a new way to write an exploit that uses the same basic method. How long, exactly, do you think it's going to take before Microsoft recognizes this and fixes the design flaws instead of releasing patches which amount to little more then sticking their finger in the crack in the dam?
Where exactly in my post did I bash Windows? I pointed out that I agree that server admins that don't patch systems are at fault if their server has problems. But, I also pointed out that we're not talking about servers being exploited, we're talking about software which runs on top of Windows consistently have gaping holes.
Microsoft could put as many firewalls as they want to into their product, but IE (or out look, or uPNP, name your exploit of the week) continually has holes in it. A web browser should not have the capability to cause the havoc that IE has caused for Microsoft users (and don't even get me started on Outlook).
Not that there isn't plenty of reasons to bash Windows, please don't imply that I'm doing something that I'm obviously not doing (and don't automatically assume that I know nothing about Windows just because I know something of Linux).
Hmm ... or maybe it's because that problem with Linux went away a long time ago. A default workstation install of Red Hat Linux 7.2 has zero open ports and a firewall that blocks access to all ports under 1024.
... there has never been an exploit like this released against Linux, there is an exploit like this released against Windows about once a month ... I think we can safely start saying it's Microsoft's fault at some point.
Now, obviously if someone sets up a server and doesn't patch, that person is an idiot (and that is true no matter what OS he/she is running). Unfortunately for your argument, we're talking about an instant messenger client and a web browser, not things that are likely to be installed on a server. The fact is, you can't exploit my Linux system via Mozilla/Konq/Galeon/Netscape, yet every other week, a new way to exploit Windows using IE pops up.
So, in conclusion, your argument is completely irrelevant to the topic at hand
What stores do you shop in? Fry's has (on the shelves) Windows 2000, 2000 Pro, 2000 Server, XP Home, XP Pro, and Me. Best Buy and Circuit City have mostly the same (they didn't have 2000 Server IIRC). Even if you don't want to count the different flavors of the same version, that is 3 different versions that are easily available at the store.
... it was called Win32. The fact that they broke the API repeatedly, and every different patch level of every different version of Windows caused new and different bugs in Win32 does not make me feel even the slightest bit confident that XP is going to make this any better.
.NET (another new platform). MS will make sure that you always need to have OS specific hooks and code paths to work reliably across the commonly used versions of Windows.
And outside of stores, there are all of the different embedded versions of Windows, etc. MS has had one platform for the past 7 years
And XP is supposed to be phased out within 2 years for the next version of Windows that will be all
If people actually went to the Mono website and read some things about it they can see that *THE MONO TEAM DOESN'T CARE IF MICROSOFT FOLLOWS THE SPEC*
C#, the CLR, and the Class Libraries *could* solve some of the problems that Miguel saw Gnome was having. Since it could solve some of these problems they are implementing the ECMA spec. If MS doesn't follow the spec, great. If they do follow the spec, great. This *does not effect Mono at all*. The project is not looking for interoperability. If they get it, it's a great benefit, but the primary goal is to solve problems that they personally have.
If you'd have read the article you'd have seen that it comes with a 40GB harddisk (for the internal bay) and 100 Mb/s ethernet adaptor, a usb keyboard and mouse, a vga output adaptor *and* the software. I don't know about you, but that looks like it could be about $200 worth of hardware alone if it's quality stuff.
So, you work on parts of Gnome, but API references and thousands of lines of code using those API's isn't enough for you? I'm sorry, but what parts of Gnome do you work on? I've worked on parts of Gnome too, and yet, somehow, I've managed. Hell, back in the Gnome 0.30 days, there was no documentation to speak of at all. And if you're complaining, why aren't you complaining about extensive documentation about libraries like Bonobo and GConf? They're certainly used by more applications then eel and gtkhtml (gtkhtml2 is a bonobo component to boot). Even some docs on just writing Nautilus Views would be alot more useful then docs on eel or librsvg.
Btw, before claiming someone should get a clue and information, make sure they don't know what they're talking about.
"The other three are documented if you download them from cvs and run gtk-doc on them."
Or not only are you wrong, but your reading comprehension skills are way below those of a first grader.
libeel is private to Nautilus. The other three are documented if you download them from cvs and run gtk-doc on them. And, none of those are core gnome libraries so you still haven't said anything about gnome core documentation.
As far as Nautilus and gnumeric, those are huge apps and porting will take a while. Of course they aren't adding features. But the code is being cleaned up as they go.
Go and troll somewhere else.
Everyone is. The day of the IPO, the stock opens on the market at that price. There have been IPO's where the stock went down on the first day. The fact that someone doesn't manage to buy the stock at that price goes up is just too bad. That's how the stock market works. If everyone got to preorder IPO shares at that price, then the price would never move on the stock, thus defeating the purpose of the stock market.
Tell me what it is you're referring about and perhaps I can respond. As far as I've seen, the biggest difference between KDE 2.2 and 3 is they're using Qt 3 now. And I can't really say that Qt 3 impresses me very much. OK, they added database aware widgets, I'll use Gnome-db.
So because gnome hasn't up'd it's version number fast enought it's not as good as KDE? (I remember everyone complaining about them bumping it up too fast back around the 1.0 release).
Gnome 2 is internationalized, has antialiased text, has a very configurable interface. The control center has been just about completely rewritten and is very slick.
If there is a UI difference between apps, complain to the app writer. But, gtk2 will make it much easier to write apps with a common look and feel and has made some nice improvements to the theme system.
Kmail is nice, Evolution is nicer, IMHO and Pan is just as good or better than KNode (again, IMHO). Glade and libglade couldn't make writing apps easier and Anjuta (especially with the work they're doing on Anjuta 2) is a very nice IDE.
If you want to think you are that much better than me for using KDE, please go ahead and do so. But your comment shows that you really are not aware of the capabilities of the current Gnome or of the huge advances that Gnome 2 has made. Things like the Pango font render, Bonobo, etc. are at the cutting edge of Linux desktop technology.
1. So go download one. There are Linux kernel debuggers available, just none in the default kernel tree.
2. Grub
3. You'll always find flaws in your design no matter how much thought is put into it. Thus the evolutionary progress of *all* OS's. OS X versus OS X.1 is a great example of that.
4. I would have to say that that number of people using and developing Linux under all the different configurations out there can be construed as taking input from "a number of external sources." Not to mention Red Hat, Mandrake, SuSE, etc. and their feedback.
5. The Linux source code is controlled by the maintainer of the kernel series. Different people have tried different source control tools and none have worked yet.
I'm not going to say there aren't things wrong with Linux, but you certainly haven't pointed any out here. And none of what you said makes it easier to debug OS X. And even if you do find a bug in it, are you going to write up a patch and submit it to the maintainers? Nope, didn't think so.